• ETRI Journal
• /
• 제33권4호
• /
• pp.600-610
• /
• 2011

This paper presents a novel approach to face detection by localizing faces as the goal-specific saliencies in a scene, using the framework of selective visual attention of a human with a particular goal in mind. The proposed approach aims at achieving human-like robustness as well as efficiency in face detection under large scene variations. The key is to establish how the specific knowledge relevant to the goal interacts with the bottom-up process of external visual stimuli for saliency detection. We propose a direct incorporation of the goal-related knowledge into the specification and/or modification of the internal process of a general bottom-up saliency detection framework. More specifically, prior knowledge of the human face, such as its size, skin color, and shape, is directly set to the window size and color signature for computing the center of difference, as well as to modify the importance weight, as a means of transforming into a goal-specific saliency detection. The experimental evaluation shows that the proposed method reaches a detection rate of 93.4% with a false positive rate of 7.1%, indicating the robustness against a wide variation of scale and rotation.

• 제어로봇시스템학회논문지
• /
• 제9권12호
• /
• pp.1009-1018
• /
• 2003

In a park or street, we can see many people Jogging or walking with their dogs chasing their masters. In the previous study, an entertainment robot, CARTRI that imitates the dog's behavior was created. The robot's task was chasing a moving goal that was recognized as the master. The physical structure of the CARTRI robot was three-wheel type locomotion system. The sensor system which could detect the position of the master in the outdoor space, was consists of a signal transmitter which was held by the master and five ultrasonic receivers which were mounted on the robot. In the experiment, the robot could chase a human walking in outdoor space like a park. But it could not avoid obstacles and its behavior was only goal-chasing behavior because of the limit of the sensor system. In this study, an improved RF-ultrasonic sensor system which can detect both goal and obstacle is developed in order to enable the CARTRI robot to carry out various behavior. The sensor system has increased angle resolution by using eight ultrasonic receivers instead of five in the previous study. And it can detect obstacle by using reflective type ultrasonic sensors. The sensor system is designed so that detection of goal and obstacle could be conducted in one sampling period. The Performance of the developed sensor system is evaluated through experiments.

• 정보처리학회논문지B
• /
• 제14B권1호
• /
• pp.1-4
• /
• 2007

본 논문에서는 축구 동영상으로부터 자동적으로 하이라이트 장면들을 검출하는 방법을 제안한다. 축구 동영상에서 하이라이트 장면들은 슈팅 장면들이나 골 장면들로 정의 될 수 있다. 우리는 축구 동영상에 대한 구조적 분석을 통해서 일반적으로 골 포스트(goal posts) 영역 주위에서 하이라이트 장면들이 나타나는 것과 하이라이트 장면 이후에는 TV 카메라가 축구 선수들이나 관중석을 확대해서 보여주는 것을 확인할 수 있었다. 본 논문에서 축구 동영상으로부터 하이라이트 장면들을 검출하기 위한 과정은 세 단계로 구성된다. 첫 번째 단계는 통계적인 문턱치(statistical threshold)를 이용한 그라운드(playing field) 영역을 추출한다. 두 번째 단계는 골 포스트를 찾기 위해서 그라운드 영역과 그라운드가 아닌 영역들의 경계선 부분을 검출한다. 그리고 마지막 단계에서는 축구 선수나 관객들의 확대 장면을 검출하기 위해서 그라운드가 아닌 영역들에 대해서 connected component labeling 기법을 적용하여 한 장면 내에서 그라운드가 아닌 영역들의 비율을 계산한다. 본 논문에서는 하이라이트 장면 검출에 대한 성능을 평가하기 위하여 정확률(precision)과 재현율(recall)을 사용하고, 실험을 통하여 제안된 방법이 정확률 95.2%, 재현율 854%로 축구 동영상에서 하이라이트 장면을 효과적으로 검출할 수 있음을 확인하였다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 추계종합학술대회 논문집(3)
• /
• pp.81-84
• /
• 2000

In this paper, we propose an efficient method to extract a goal area which may be closely related to the scoring highlight. In our method, the boundary between the ground and the non-ground area is used. An efficient methods for a rapid detection of both the boundary and then the goal area are proposed. Our simulation results show that our method is very reliable and takes less processing time compared with previous methods. This performance improvements may be caused by the use of a general simple feature.

• International Journal of Computer Science & Network Security
• /
• 제22권11호
• /
• pp.77-86
• /
• 2022

The destructive impact of contagious diseases outbreaks on all life facets necessitates developing effective solutions to control these diseases outbreaks. This research proposes an end-to-end, data-driven platform which consists of multiple modules that are working in harmony to achieve a concrete goal: early detection of contagious diseases outbreaks (i.e., epidemic diseases detection). Achieving that goal enables decision makers and people in power to act promptly, resulting in robust prevention management of contagious diseases. It must be clear that the goal of this proposed platform is not to predict or forecast the spread of contagious diseases, rather, its goal is to promptly detect contagious diseases outbreaks as they happen. The front end of the proposed platform is a web-based dashboard that visualizes diseases outbreaks in real-time on a real map. These outbreaks are detected via another component of the platform which utilizes data mining techniques and algorithms on gathered datasets. Those gathered datasets are managed by yet another component. Specifically, a mobile application will be the main source of data to the platform. Being a vital component of the platform, the datasets are managed by a DBMS that is specifically tailored for this platform. Preliminary results are presented to showcase the performance of a prototype of the proposed platform.

• 경영과정보연구
• /
• 제22권
• /
• pp.211-229
• /
• 2007

The ISO/TS16949 APQP goal of defect prevention and decrease of spread waste, is the customer satisfaction which leads a continuous improvement and profit creation. The quality expense where the most is caused by but with increase of production initial quality problem occurrence is increasing to is actuality. Like this confirmation amendment. with the problem which is forecast in the place development at the initial stage which it does completeness it does not confront not to be able, production phase to be imminent, the problem accumulates and it talks the development shedding of which occurs. In opposition, prediction confrontation. is forecast in development early stage to and it is a structure which does not occur a problem to production early stage. Like this development is a possibility of accomplishing competitive company from production phase. Which attains an goal of, chance cause it leads a APQP activity (common cause) with special cause prevention & detection the connection characteristic of the focus technique against a interaction is important. And the customer requirement satisfaction and must convert a APQP goal of attainment at the key characteristics action step. (1) The Prevention - with Design FMEA application prevention of the present design management/detection, (2) the Detection (prevention/detection) - with Process FMEA application prevention of the present process control/detection, (3) Special Cause - statistical process control (SPC) 4M cause spread removal, (4) Common Cause - statistical process control (SPC) the nothing zero defect which leads the continuous improvement back of spread with application it will be able to attain with application.

• 한국신뢰성학회지:신뢰성응용연구
• /
• 제14권1호
• /
• pp.37-43
• /
• 2014

This paper deals with a design of observer based fault detection filter for a boiler-turbine control system. The goal is to present a method for rapid sensor fault detection in order to enhance the reliability of boiler-turbine operation in the thermal power plant. Our fault detection filter can be designed via solutions of linear matrix inequalities. In order to demonstrate the efficacy of our design method, numerical simulations are provided.

• Smart Structures and Systems
• /
• 제15권2호
• /
• pp.315-334
• /
• 2015

A main goal of this study is to propose a damage detection technique to detect and localize damages of a top-tensioned riser. In this paper, the top-tensioned finite element (FE) model is considered as an analytical model of the riser, and a vibration-based damage detection method is proposed. The present method consists of a FE model updating and damage index method. In order to accomplish the goal of this study, first, a sensitivity-based FE model updating method using natural frequencies and zero frequencies is introduced. Second, natural frequencies and zero frequencies of the axial mode on the top-tensioned riser are estimated by eigenvalue analysis. Finally, the locations and severities of the damages are estimated from the damage index method. Three numerical examples are considered to verify the performance of the proposed method.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2021년도 추계학술대회
• /
• pp.422-424
• /
• 2021

In this paper, it leads the approach of fusing multiple RGB cameras for visual objects recognition based on deep learning with convolution neural network and 3D Light Detection and Ranging (LiDAR) to observe the environment and match into a 3D world in estimating the distance and position in a form of point cloud map. The goal of perception in multiple cameras are to extract the crucial static and dynamic objects around the autonomous vehicle, especially the blind spot which assists the AV to navigate according to the goal. Numerous cameras with object detection might tend slow-going the computer process in real-time. The computer vision convolution neural network algorithm to use for eradicating this problem use must suitable also to the capacity of the hardware. The localization of classified detected objects comes from the bases of a 3D point cloud environment. But first, the LiDAR point cloud data undergo parsing, and the used algorithm is based on the 3D Euclidean clustering method which gives an accurate on localizing the objects. We evaluated the method using our dataset that comes from VLP-16 and multiple cameras and the results show the completion of the method and multi-sensor fusion strategy.

• 한국멀티미디어학회논문지
• /
• 제23권8호
• /
• pp.915-926
• /
• 2020

Outlier detection methods help one to identify unusual instances in data that may correspond to erroneous, exceptional, or surprising events or behaviors. This work studies conditional outlier detection, a special instance of the outlier detection problem, in the context of incorrect data label identification. Unlike conventional (unconditional) outlier detection methods that seek abnormalities across all data attributes, conditional outlier detection assumes data are given in pairs of input (condition) and output (response or label). Accordingly, the goal of conditional outlier detection is to identify incorrect or unusual output assignments considering their input as condition. As a solution to conditional outlier detection, this paper proposes the ratio-based outlier scoring (ROS) approach and its variant. The propose solutions work by adopting conventional outlier scores and are able to apply them to identify conditional outliers in data. Experiments on synthetic and real-world image datasets are conducted to demonstrate the benefits and advantages of the proposed approaches.